INDEX

Explanations

names and job titles of individuals

oai_token-act-pair · gpt-3.5-turbo

proper nouns related to people and organizations

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

]=

-0.69

OULD

-0.67

ORS

-0.60

ModLoader

-0.60

 toget

-0.59

 accordingly

-0.59

raq

-0.58

 ACTIONS

-0.58

 thereto

-0.58

Around

-0.57

POSITIVE LOGITS

 whose

0.90

quartered

0.80

0.76

 whereas

0.71

.;

0.67

but

0.65

0.64

and

0.63

 whom

0.61

who

0.61

Activations Density 0.884%

names and job titles of individuals

proper nouns related to people and organizations

No Comments

No Known Activations