INDEX

Explanations

proper nouns or names of individuals

oai_token-act-pair · gpt-3.5-turbo

references to specific individuals, particularly those with the surname "Freeman"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

assic

-0.91

ificial

-0.88

ģĸ

-0.80

oiler

-0.78

rolog

-0.77

ategory

-0.76

aron

-0.76

ãĤ°

-0.75

essee

-0.73

prus

-0.71

POSITIVE LOGITS

 Freeman

0.79

 Scouting

0.71

 Scrolls

0.69

 Duff

0.69

 Cheong

0.67

ufact

0.67

iquette

0.66

boro

0.66

 Morales

0.65

lings

0.65

Activations Density 0.027%

proper nouns or names of individuals

references to specific individuals, particularly those with the surname "Freeman"

No Comments

No Known Activations