INDEX

Explanations

comparisons between different quantities or values

oai_token-act-pair · gpt-3.5-turbo

the word "compared" to highlight comparisons between different statistics or data points

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

eland

-0.74

vous

-0.72

stall

-0.71

adia

-0.70

authorized

-0.66

hold

-0.65

fruit

-0.65

OTA

-0.63

 recess

-0.63

cot

-0.63

POSITIVE LOGITS

 favorably

0.89

 guiActiveUn

0.76

 objectively

0.75

lihood

0.71

Compare

0.69

 Compare

0.68

 comparing

0.64

 contemporaries

0.63

uple

0.62

 dishon

0.62

Activations Density 0.016%

comparisons between different quantities or values

the word "compared" to highlight comparisons between different statistics or data points

No Comments

No Known Activations