INDEX

Explanations

phrases with a negative connotation indicating problems or challenges

oai_token-act-pair · gpt-3.5-turbo

phrases that express negation or the concept of "not a."

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Edit

-0.80

 nonetheless

-0.74

Apparently

-0.71

çīĪ

-0.69

Links

-0.68

unknown

-0.67

edit

-0.66

Anim

-0.65

unders

-0.64

hours

-0.64

POSITIVE LOGITS

 coincidence

1.20

 anymore

0.99

 mere

0.91

 pleasant

0.90

 slightest

0.89

verse

0.88

 necessarily

0.87

nor

0.85

 bother

0.84

 nuisance

0.83

Activations Density 0.093%

phrases with a negative connotation indicating problems or challenges

phrases that express negation or the concept of "not a."

No Comments

No Known Activations

phrases with a negative connotation indicating problems or challenges

phrases that express negation or the concept of "not a."

No Comments

No Known Activations