INDEX

Explanations

phrases related to arguments or presenting a persuasive case

oai_token-act-pair · gpt-3.5-turbo

instances of the phrase "make a case" or variations of it

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

attery

-0.76

autical

-0.69

orks

-0.69

onder

-0.69

elta

-0.68

atomic

-0.68

etsk

-0.67

pez

-0.67

ummer

-0.66

CG

-0.64

POSITIVE LOGITS

cases

0.87

Against

0.80

 against

0.79

case

0.76

backs

0.74

BACK

0.71

 Keen

0.70

 convinc

0.67

 Against

0.65

why

0.64

Activations Density 0.033%

phrases related to arguments or presenting a persuasive case

instances of the phrase "make a case" or variations of it

No Comments

No Known Activations