INDEX

Explanations

phrases suggesting hypothetical situations with specific actions or consequences

oai_token-act-pair · gpt-3.5-turbo

conditional statements and hypothetical scenarios

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

TION

-0.66

ahead

-0.61

Enough

-0.59

Cook

-0.58

 Brill

-0.58

Ready

-0.58

sis

-0.57

 Ahead

-0.57

Tok

-0.57

 contained

-0.57

POSITIVE LOGITS

 expect

1.31

 imagine

1.29

 suppose

1.19

 assume

1.13

 presume

1.12

 wonder

1.03

agine

0.99

 speculate

0.99

 think

0.97

 argue

0.96

Activations Density 0.072%

phrases suggesting hypothetical situations with specific actions or consequences

conditional statements and hypothetical scenarios

No Comments

No Known Activations