INDEX

Explanations

references or mentions of specific terms, concepts, or items

oai_token-act-pair · gpt-3.5-turbo

mentions of references and citations

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

otos

-0.74

asus

-0.66

anooga

-0.66

ãĥ£

-0.66

sylv

-0.66

awar

-0.65

ihad

-0.65

oly

-0.65

wich

-0.64

ilus

-0.62

POSITIVE LOGITS

ibly

0.81

ENCE

0.79

 counting

0.78

 references

0.78

 thereto

0.77

 therein

0.76

ibility

0.75

ENC

0.74

itiveness

0.73

count

0.73

Activations Density 0.036%

references or mentions of specific terms, concepts, or items

mentions of references and citations

No Comments

No Known Activations