INDEX

Explanations

phrases related to revealing or disclosing something

oai_token-act-pair · gpt-3.5-turbo

phrases indicating emergence or departure from a situation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 apologizing

-0.76

 estimating

-0.70

OUP

-0.69

 answ

-0.67

 anticipating

-0.67

amera

-0.66

 employing

-0.63

 recalling

-0.63

 referring

-0.63

 lineback

-0.63

POSITIVE LOGITS

aband

0.78

natureconservancy

0.78

iped

0.75

edge

0.73

unks

0.72

blers

0.68

redients

0.67

fits

0.67

packs

0.65

 footsteps

0.64

Activations Density 0.675%

phrases related to revealing or disclosing something

phrases indicating emergence or departure from a situation

No Comments

No Known Activations