INDEX

Explanations

words related to reactions or responses from people or groups

oai_token-act-pair · gpt-3.5-turbo

terms related to reactions and responses

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

skip

-0.66

Zot

-0.65

bilt

-0.63

atan

-0.63

cutting

-0.63

ramer

-0.61

cut

-0.61

BALL

-0.60

sonian

-0.59

rey

-0.59

POSITIVE LOGITS

aries

1.13

ivated

1.04

 thereto

0.98

ivating

0.91

ivation

0.85

ariat

0.83

iveness

0.83

naires

0.79

aire

0.79

gif

0.76

Activations Density 0.074%

words related to reactions or responses from people or groups

terms related to reactions and responses

No Comments

No Known Activations