INDEX

Explanations

names of specific individuals

oai_token-act-pair · gpt-3.5-turbo

names of prominent figures and organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

awa

-0.61

 Defin

-0.59

OPLE

-0.59

è¦ļéĨĴ

-0.58

 curve

-0.58

©¶æ

-0.57

 gradient

-0.57

 Democr

-0.57

 rainbow

-0.56

fog

-0.55

POSITIVE LOGITS

olver

0.67

etc

0.66

agen

0.63

)'

0.63

awan

0.63

guard

0.62

oshenko

0.61

rup

0.61

sat

0.60

avia

0.60

Activations Density 0.441%

names of specific individuals

names of prominent figures and organizations

No Comments

No Known Activations