INDEX

Explanations

terms related to online content or web links

oai_token-act-pair · gpt-3.5-turbo

references to specific abbreviations or codes, particularly CL (likely standing for something like "Classification Level" or similar)

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

ãĥĦ

-0.93

tale

-0.87

fitting

-0.85

ãĤ®

-0.83

fit

-0.76

ãĥĥ

-0.76

hide

-0.75

fal

-0.75

sov

-0.75

spect

-0.74

POSITIVE LOGITS

OSED

1.13

INTON

1.09

OCK

0.98

IENT

0.95

OTH

0.89

isters

0.86

opez

0.85

avier

0.85

OVER

0.84

AMP

0.83

Activations Density 0.006%

terms related to online content or web links

references to specific abbreviations or codes, particularly CL (likely standing for something like "Classification Level" or similar)

No Comments

No Known Activations