INDEX

Explanations

phrases mentioning specific individuals or groups

oai_token-act-pair · gpt-3.5-turbo

the word "them" in various contexts

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Deal

-0.75

Press

-0.69

Jo

-0.66

order

-0.65

Chain

-0.65

ILE

-0.65

deal

-0.64

 Patton

-0.63

Rush

-0.63

Monster

-0.62

POSITIVE LOGITS

 selves

1.14

atically

1.00

atic

0.99

selves

0.87

 conduc

0.81

 outwe

0.75

self

0.70

 sinks

0.70

 succeeded

0.70

atics

0.69

Activations Density 0.038%

phrases mentioning specific individuals or groups

the word "them" in various contexts

No Comments

No Known Activations