INDEX

Explanations

phrases indicating a decision or choice being made to proceed with something

oai_token-act-pair · gpt-3.5-turbo

the phrase "go with" and its variations used 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

clips

-0.65

stem

-0.65

cent

-0.63

esley

-0.61

blast

-0.59

former

-0.59

pring

-0.59

wordpress

-0.59

ilight

-0.59

println

-0.58

POSITIVE LOGITS

 regards

1.25

 regard

1.14

 impunity

1.04

drawn

0.91

 respect

0.91

standing

0.89

draw

0.89

 confidence

0.82

 utmost

0.81

resses

0.75

Activations Density 0.110%

phrases indicating a decision or choice being made to proceed with something

the phrase "go with" and its variations used in various contexts

No Comments

No Known Activations