INDEX

Explanations

words related to glasses or glass objects

oai_token-act-pair · gpt-3.5-turbo

references to glass and its various forms and uses

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

CVE

-0.71

ilitary

-0.68

IRED

-0.67

enance

-0.64

Filename

-0.63

avez

-0.61

Ake

-0.61

ften

-0.60

 sanction

-0.60

unct

-0.59

POSITIVE LOGITS

 shards

1.03

gow

1.01

 beads

0.99

works

0.96

 bottles

0.94

 pane

0.94

glass

0.93

loo

0.92

caps

0.92

acles

0.91

Activations Density 0.021%

words related to glasses or glass objects

references to glass and its various forms and uses

No Comments

No Known Activations