INDEX

Explanations

phrases related to indicating direction or focus

oai_token-act-pair · gpt-3.5-turbo

references to guidance or direction

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

 distraction

-0.62

ornia

-0.58

ment

-0.57

 Crush

-0.57

 headlines

-0.56

 cele

-0.55

ãĥ¡

-0.55

 Mansion

-0.55

GHC

-0.53

 Aram

-0.53

POSITIVE LOGITS

oward

0.86

heit

0.86

ggle

0.80

geon

0.77

 forth

0.76

athe

0.72

ysc

0.71

onge

0.70

arily

0.69

ugh

0.68

Activations Density 0.385%

phrases related to indicating direction or focus

references to guidance or direction

No Comments

No Known Activations