INDEX

Explanations

phrases indicating personal opinions, beliefs, or thoughts

oai_token-act-pair · gpt-3.5-turbo

pronouns, particularly those referring to people, in context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

odder

-0.65

 Hicks

-0.64

ardon

-0.63

epad

-0.61

 Bots

-0.59

 Cunningham

-0.59

İĭ

-0.59

OTOS

-0.57

inational

-0.57

 Bland

-0.57

POSITIVE LOGITS

 fared

0.91

 handled

0.81

'd

0.81

've

0.81

 fucked

0.80

 stacked

0.78

 cope

0.76

 relates

0.76

 handle

0.75

 behave

0.75

Activations Density 0.113%

phrases indicating personal opinions, beliefs, or thoughts

pronouns, particularly those referring to people, in context

No Comments

No Known Activations