INDEX

Explanations

proper nouns, likely related to places or names

oai_token-act-pair · gpt-3.5-turbo

names and terms associated with specific entities or brands

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 Stan

-0.76

 Percy

-0.72

Stan

-0.72

 Westbrook

-0.70

 Clover

-0.69

 Vert

-0.69

 PRES

-0.69

 Arthur

-0.69

 buck

-0.68

POSITIVE LOGITS

1.60

ii

1.20

ih

1.11

iak

1.11

iy

1.11

iot

1.10

iom

1.08

ici

1.08

ni

1.07

iki

1.06

Activations Density 0.154%

proper nouns, likely related to places or names

names and terms associated with specific entities or brands

No Comments

No Known Activations