INDEX

Explanations

words related to recovery or returning to a previous state

oai_token-act-pair · gpt-3.5-turbo

occurrences of the prefix "rec" indicating reoccurring or repeated actions or conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

MENTS

-0.79

loo

-0.75

nuts

-0.73

NESS

-0.72

MENT

-0.72

 Gemini

-0.69

externalActionCode

-0.69

 Nest

-0.68

OPLE

-0.67

hole

-0.67

POSITIVE LOGITS

ourses

1.16

ievers

1.08

onduct

1.08

ategor

1.06

ieving

1.04

uters

1.03

ursion

1.03

urrence

1.02

usal

1.02

iever

1.01

Activations Density 0.010%

words related to recovery or returning to a previous state

occurrences of the prefix "rec" indicating reoccurring or repeated actions or conditions

No Comments

No Known Activations