INDEX

Explanations

verbs related to demonstrating or showing evidence of something

oai_token-act-pair · gpt-3.5-turbo

instances of the word "prove" 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 @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 contrace

-0.72

 newsletters

-0.69

artment

-0.69

itch

-0.67

 newsp

-0.64

roman

-0.63

phrine

-0.63

letal

-0.62

 column

-0.62

atches

-0.62

POSITIVE LOGITS

 proves

0.87

 prove

0.84

 proved

0.79

 proving

0.76

URE

0.73

rers

0.71

 dispro

0.70

\\\\\\\\

0.70

ELY

0.69

INGS

0.67

Activations Density 0.014%

verbs related to demonstrating or showing evidence of something

instances of the word "prove" in various contexts

No Comments

No Known Activations