INDEX

Explanations

references to a specific publication named "Globe"

oai_token-act-pair · gpt-3.5-turbo

references to the term "Globe" in various contexts, likely related to news or media

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

ivated

-0.82

ramid

-0.82

esson

-0.82

amac

-0.79

dj

-0.75

sed

-0.74

tan

-0.74

hyde

-0.73

sen

-0.72

unta

-0.72

POSITIVE LOGITS

 Globe

1.07

 Newspaper

0.91

 Herald

0.88

 Scotia

0.78

 Harbour

0.75

 Harbor

0.74

 Observer

0.72

 Newsp

0.69

 Wire

0.69

 Shards

0.67

Activations Density 0.004%

references to a specific publication named "Globe"

references to the term "Globe" in various contexts, likely related to news or media

No Comments

No Known Activations