INDEX

Explanations

phrases related to providing explanations or reasons

oai_token-act-pair · gpt-3.5-turbo

phrases indicating causation or justification

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

 havoc

-0.73

acters

-0.65

 Chains

-0.65

 Creatures

-0.65

ahime

-0.62

bill

-0.62

aths

-0.61

nan

-0.60

 adolesc

-0.59

iage

-0.57

POSITIVE LOGITS

 regarding

0.84

 concerning

0.75

 glean

0.74

 stating

0.73

uristic

0.69

 whereby

0.69

 summar

0.66

bara

0.66

©¶æ

0.66

about

0.65

Activations Density 0.494%

phrases related to providing explanations or reasons

phrases indicating causation or justification

No Comments

No Known Activations