INDEX

Explanations

verbs or phrases related to analyzing or interpreting information

oai_token-act-pair · gpt-3.5-turbo

prepositions and their relationship to meaning-making in context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 teasp

-0.95

osponsors

-0.81

starter

-0.77

Â¯Â¯Â¯Â¯

-0.76

 pione

-0.75

 GOODMAN

-0.74

atsuki

-0.74

idential

-0.73

encia

-0.72

ę

-0.72

POSITIVE LOGITS

 notions

1.04

 phrases

1.00

 whatever

0.96

 everything

0.95

 myths

0.93

 what

0.92

 tales

0.92

 perceptions

0.90

 anecdotes

0.88

 things

0.88

Activations Density 0.329%

verbs or phrases related to analyzing or interpreting information

prepositions and their relationship to meaning-making in context

No Comments

No Known Activations