INDEX

Explanations

phrases related to specific locations or entities

oai_token-act-pair · gpt-3.5-turbo

the presence of specific web-related terms or identifiers

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

IFrame

Link

Negative Logits

STD

-0.82

-+-+

-0.82

actionDate

-0.82

ICLE

-0.81

-0.77

¿½

-0.74

odi

-0.73

INTON

-0.73

STAR

-0.73

 referen

-0.72

POSITIVE LOGITS

 Revis

0.88

 Falls

0.81

 Manor

0.76

phia

0.75

 Chronicles

0.71

 County

0.71

 Unified

0.70

 Chronicle

0.70

hire

0.69

 Parish

0.69

Activations Density 0.415%

phrases related to specific locations or entities

the presence of specific web-related terms or identifiers

No Known Activations