INDEX

Explanations

information on various news topics and updates

oai_token-act-pair · gpt-3.5-turbo

ellipses or omissions that suggest continuity in a narrative

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

heit

-0.79

schild

-0.74

umenthal

-0.72

geist

-0.64

kowski

-0.63

pread

-0.62

igslist

-0.62

wagon

-0.62

metics

-0.61

bara

-0.59

POSITIVE LOGITS

 MUST

1.26

thumbnails

0.71

must

0.71

Roberts

0.66

 Became

0.64

Hide

0.63

Must

0.62

lect

0.61

Spawn

0.60

ãĥ¤

0.59

Activations Density 0.015%

information on various news topics and updates

ellipses or omissions that suggest continuity in a narrative

No Comments

No Known Activations