INDEX

Explanations

names of specific individuals or entities

oai_token-act-pair · gpt-3.5-turbo

names of key individuals associated with specific events or narratives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ordinary

-0.77

print

-0.77

rowth

-0.65

ambers

-0.65

Band

-0.65

NX

-0.65

Quote

-0.63

intendent

-0.62

hens

-0.62

 incre

-0.61

POSITIVE LOGITS

 Saul

1.05

 Berman

0.92

Nab

0.76

 Goodman

0.76

 Faul

0.72

scripts

0.71

 Judah

0.71

tering

0.71

kj

0.70

 Boone

0.70

Activations Density 0.028%

names of specific individuals or entities

names of key individuals associated with specific events or narratives

No Comments

No Known Activations