INDEX

Explanations

instances of specific coded abbreviations

oai_token-act-pair · gpt-3.5-turbo

references to the Diagnostic and Statistical Manual of Mental Disorders (DSM)

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

swick

-0.86

============

-0.71

 symmetry

-0.70

Reviewer

-0.68

ishment

-0.64

issance

-0.64

========

-0.64

ufact

-0.63

 pains

-0.63

ished

-0.61

POSITIVE LOGITS

omething

1.13

IFF

0.96

OME

0.91

LR

0.87

yll

0.87

FW

0.86

DK

0.86

HI

0.85

CAP

0.85

III

0.85

Activations Density 0.035%

instances of specific coded abbreviations

references to the Diagnostic and Statistical Manual of Mental Disorders (DSM)

No Comments

No Known Activations