INDEX

Explanations

comparisons of different groups or entities

oai_token-act-pair · gpt-3.5-turbo

statistics related to health and mortality rates

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

 wrench

-0.66

 miracles

-0.61

 Korra

-0.59

 leverage

-0.58

 symbolism

-0.57

awar

-0.56

 translator

-0.55

 sails

-0.55

 someday

-0.55

 mediation

-0.54

POSITIVE LOGITS

dL

0.87

average

0.84

amphetamine

0.83

 peers

0.78

icient

0.75

 counterparts

0.74

normal

0.72

rats

0.71

ffee

0.71

ustomed

0.68

Activations Density 0.238%

comparisons of different groups or entities

statistics related to health and mortality rates

No Comments

No Known Activations