INDEX

Explanations

quotations from individuals being reported on by the media

oai_token-act-pair · gpt-3.5-turbo

references to media interactions or statements made to reporters

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ez

-0.72

osate

-0.69

UTH

-0.65

istries

-0.63

gradation

-0.61

omorphic

-0.61

STD

-0.61

Quantity

-0.61

emonic

-0.59

ological

-0.59

POSITIVE LOGITS

 gathered

0.97

 aboard

0.90

 backstage

0.88

 afterward

0.88

 assembled

0.87

 stationed

0.84

hips

0.80

 reporters

0.79

 yesterday

0.78

 Thursday

0.78

Activations Density 0.047%

quotations from individuals being reported on by the media

references to media interactions or statements made to reporters

No Comments

No Known Activations