INDEX

Explanations

statements or phrases indicating a particular perspective or viewpoint

oai_token-act-pair · gpt-3.5-turbo

expressions describing perspectives or interpretations of situations

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

usters

-0.82

erville

-0.77

uster

-0.75

 livest

-0.74

ividual

-0.65

 recomm

-0.65

 sugg

-0.65

è¦ļéĨĴ

-0.62

 subsequ

-0.62

 blockers

-0.61

POSITIVE LOGITS

fare

1.06

finding

0.88

forward

0.83

bill

0.75

ward

0.72

point

0.71

 Dolphin

0.70

finder

0.69

ever

0.66

CHO

0.65

Activations Density 0.030%

statements or phrases indicating a particular perspective or viewpoint

expressions describing perspectives or interpretations of situations

No Comments

No Known Activations