INDEX

Explanations

phrases related to expressing opinions or explanations

oai_token-act-pair · gpt-3.5-turbo

the phrase "that's what" in various contexts

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

robe

-0.81

uttering

-0.63

ãĥ¼ãĥ³

-0.63

eer

-0.62

fw

-0.57

jj

-0.57

holding

-0.57

case

-0.56

UNE

-0.56

still

-0.55

POSITIVE LOGITS

 happens

1.14

 happened

1.07

soever

1.06

 happ

0.96

 separates

0.84

 sorts

0.82

 transpired

0.78

 kinds

0.78

 distinguishes

0.76

itionally

0.74

Activations Density 0.049%

phrases related to expressing opinions or explanations

the phrase "that's what" in various contexts

No Comments

No Known Activations