INDEX

Explanations

legal and political terms related to actions or decisions

oai_token-act-pair · gpt-3.5-turbo

references to recent reports or decisions in a political context

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Anyway

-0.77

Zip

-0.75

 solves

-0.73

 denotes

-0.68

 adjusts

-0.68

 Anyway

-0.67

 Alright

-0.66

 doesnt

-0.65

 survives

-0.65

 behaves

-0.65

POSITIVE LOGITS

 controversial

0.91

 leaked

0.88

 purported

0.81

 alleged

0.80

 whistleblower

0.76

 dossier

0.76

 controversy

0.75

 proposed

0.75

 ambush

0.72

 proposal

0.72

Activations Density 0.455%

legal and political terms related to actions or decisions

references to recent reports or decisions in a political context

No Comments

No Known Activations