INDEX

Explanations

commands or instructions to relax and stop worrying

oai_token-act-pair · gpt-3.5-turbo

punctuations and pauses indicative of conversational or narrative breaks

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

herent

-0.73

izon

-0.67

 Impl

-0.64

vre

-0.64

amation

-0.63

isi

-0.55

DX

-0.54

è£ħ

-0.54

ãĥ¯

-0.54

anium

-0.54

POSITIVE LOGITS

 lest

0.98

 preferably

0.88

 albeit

0.82

 relax

0.70

 knowing

0.68

 innovate

0.65

Finish

0.65

 however

0.65

 though

0.64

izont

0.64

Activations Density 0.378%

commands or instructions to relax and stop worrying

punctuations and pauses indicative of conversational or narrative breaks

No Comments

No Known Activations