INDEX

Explanations

phrases related to instructions or procedures

oai_token-act-pair · gpt-3.5-turbo

affirmative statements about people's actions or state of being

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

igraph

-0.65

icy

-0.64

ESE

-0.62

Discuss

-0.62

ilts

-0.60

 airs

-0.60

)]

-0.59

ACP

-0.59

 originate

-0.58

uca

-0.57

POSITIVE LOGITS

 yourself

0.90

 yourselves

0.89

 probably

0.85

Tube

0.84

 doubtless

0.82

 Yourself

0.80

 guessed

0.80

probably

0.80

 mileage

0.79

 forgiven

0.77

Activations Density 0.237%

phrases related to instructions or procedures

affirmative statements about people's actions or state of being

No Comments

No Known Activations