INDEX

Explanations

phrases related to different methods or strategies

oai_token-act-pair · gpt-3.5-turbo

phrases that signify methods or approaches to achieve a goal

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

usters

-0.93

ĸļ

-0.89

uster

-0.77

noxious

-0.72

icio

-0.69

arthed

-0.68

 resent

-0.65

 disappoint

-0.64

omore

-0.63

asts

-0.63

POSITIVE LOGITS

finding

1.13

fare

1.03

point

1.00

forward

0.91

ward

0.91

points

0.86

 forward

0.85

finder

0.79

kell

0.78

NE

0.74

Activations Density 0.048%

phrases related to different methods or strategies

phrases that signify methods or approaches to achieve a goal

No Comments

No Known Activations