INDEX

Explanations

words or parts of words related to extension or expansion

oai_token-act-pair · gpt-3.5-turbo

variations of the prefix "ext" and associated terms

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

hari

-0.85

Downloadha

-0.77

ker

-0.77

folk

-0.75

maid

-0.74

 sharks

-0.69

UGH

-0.68

hare

-0.67

 Bard

-0.67

hart

-0.66

POSITIVE LOGITS

racted

1.56

raction

1.46

ensions

1.45

ension

1.42

ract

1.41

ortion

1.36

inct

1.34

ractor

1.26

ended

1.23

ractions

1.23

Activations Density 0.010%

words or parts of words related to extension or expansion

variations of the prefix "ext" and associated terms

No Comments

No Known Activations