INDEX

Explanations

phrases indicating uncertainty or vagueness

oai_token-act-pair · gpt-3.5-turbo

vague references to unspecified objects or concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

arest

-0.64

adapt

-0.62

raid

-0.61

ENDED

-0.58

fw

-0.58

selves

-0.57

Ã©n

-0.57

ruck

-0.56

schild

-0.56

pload

-0.55

POSITIVE LOGITS

 else

1.22

 thereof

1.10

 alike

1.03

 similar

0.99

 like

0.86

 analogous

0.86

 fancy

0.86

Else

0.84

abouts

0.82

 nonsense

0.81

Activations Density 0.084%

phrases indicating uncertainty or vagueness

vague references to unspecified objects or concepts

No Comments

No Known Activations