INDEX

Explanations

statements and opinions, often introduced with verbs like 'say', 'added', 'argued', 'claimed', 'went', 'said', 'contended', or 'noted'

oai_token-act-pair · gpt-3.5-turbo

verbs and phrases related to statements and claims

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

prus

-0.63

adesh

-0.58

ptives

-0.56

 confir

-0.53

speech

-0.52

ainted

-0.49

estern

-0.49

NL

-0.49

reporting

-0.49

 teasp

-0.48

POSITIVE LOGITS

0.89

,,

0.82

*,

0.81

,[

0.75

.,

0.73

!,

0.69

,—

0.67

,...

0.67

Â®,

0.65

,-

0.64

Activations Density 0.192%

statements and opinions, often introduced with verbs like 'say', 'added', 'argued', 'claimed', 'went', 'said', 'contended', or 'noted'

verbs and phrases related to statements and claims

No Comments

No Known Activations