INDEX

Explanations

words related to initiation or starting

oai_token-act-pair · gpt-3.5-turbo

words that indicate the start of actions or processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

raid

-0.73

verbs

-0.73

oted

-0.73

*/(

-0.72

rome

-0.72

uyomi

-0.71

rats

-0.71

orie

-0.70

pled

-0.69

iliary

-0.69

POSITIVE LOGITS

 anew

0.92

 attRot

0.81

ITIES

0.80

lihood

0.74

 "$:/

0.73

OPLE

0.72

nings

0.71

Ò

0.68

EGIN

0.68

 Valiant

0.67

Activations Density 0.017%

words related to initiation or starting

words that indicate the start of actions or processes

No Comments

No Known Activations