INDEX

Explanations

names with associated statements or questions

oai_token-act-pair · gpt-3.5-turbo

instances of dialogue or conversational exchanges

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

ascus

-0.81

anmar

-0.71

eatures

-0.67

inement

-0.66

ometown

-0.66

abad

-0.65

 unification

-0.63

 administ

-0.63

 agric

-0.62

 bride

-0.61

POSITIVE LOGITS

Yeah

1.16

Huh

1.14

Hmm

1.09

 Alright

1.07

Exactly

1.06

Hmm

1.05

 Exactly

1.05

Oh

1.04

Uh

1.04

Yes

1.02

Activations Density 0.167%

names with associated statements or questions

instances of dialogue or conversational exchanges

No Comments

No Known Activations