INDEX

Explanations

legislative and political terms

oai_token-act-pair · gpt-3.5-turbo

references to legislative actions and committee activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

Afee

-0.82

ãĤ¨ãĥ«

-0.79

obar

-0.72

ãĤ¼

-0.70

 IMAGES

-0.68

 compan

-0.67

 imagery

-0.67

PLIED

-0.66

 gradient

-0.66

 fertilizer

-0.65

POSITIVE LOGITS

 unanimously

1.17

 convened

1.14

 Subcommittee

1.11

 hearings

1.08

 recess

1.07

 adjourn

1.06

 voted

1.05

 Caucus

1.04

 subcommittee

1.04

 Majority

1.03

Activations Density 0.559%

legislative and political terms

references to legislative actions and committee activities

No Comments

No Known Activations