INDEX

Explanations

article titles or headlines containing specific phrases or names

oai_token-act-pair · gpt-3.5-turbo

references to specific individuals or entities, particularly regarding their actions or statements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ĵĺ

-0.64

 pasture

-0.64

 Stain

-0.63

 gestation

-0.62

Fn

-0.61

 Dele

-0.60

GBT

-0.58

Runner

-0.58

ADRA

-0.57

 Sussex

-0.57

POSITIVE LOGITS

etus

0.79

ilver

0.73

agi

0.71

cious

0.71

Äĩ

0.69

imgur

0.68

eper

0.67

boa

0.66

ewski

0.66

lication

0.64

Activations Density 0.360%

article titles or headlines containing specific phrases or names

references to specific individuals or entities, particularly regarding their actions or statements

No Comments

No Known Activations