INDEX

Explanations

mentions of reputations, especially those with a high level of significance or impact

oai_token-act-pair · gpt-3.5-turbo

phrases related to reputation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

tein

-0.77

cise

-0.72

pheus

-0.69

onal

-0.67

WORK

-0.66

ECK

-0.66

nes

-0.63

wed

-0.63

err

-0.62

sight

-0.61

POSITIVE LOGITS

 tarn

0.95

 reputation

0.90

 bearer

0.84

 tremend

0.82

 forged

0.81

ously

0.76

 abroad

0.76

worthiness

0.75

arily

0.75

 bestowed

0.74

Activations Density 0.089%

mentions of reputations, especially those with a high level of significance or impact

phrases related to reputation

No Comments

No Known Activations