INDEX

Explanations

phrases related to technical errors or issues

oai_token-act-pair · gpt-3.5-turbo

occurrences of error messages or indicators of issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

amous

-0.72

 savage

-0.65

 horr

-0.62

enance

-0.61

onyms

-0.61

aughter

-0.61

 depl

-0.60

kers

-0.59

vered

-0.58

 abhor

-0.58

POSITIVE LOGITS

 Please

1.08

 Reload

0.91

Please

0.78

 Something

0.76

 Responsibility

0.74

 Sorry

0.73

Typ

0.73

 Connection

0.72

 Click

0.72

 Close

0.72

Activations Density 0.025%

phrases related to technical errors or issues

occurrences of error messages or indicators of issues

No Comments

No Known Activations