INDEX

Explanations

words related to repetition or sequential events

oai_token-act-pair · gpt-3.5-turbo

recurrent themes of time and repeated events

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

obal

-0.77

jon

-0.72

onom

-0.72

UTH

-0.71

anth

-0.69

ãĥ¤

-0.69

Marginal

-0.67

nos

-0.65

Exit

-0.65

sth

-0.64

POSITIVE LOGITS

 bombard

0.80

ciating

0.74

 endlessly

0.70

 alike

0.66

 successive

0.65

=-=-=-=-

0.65

 swat

0.64

 consecut

0.63

phia

0.62

 imaginable

0.62

Activations Density 0.071%

words related to repetition or sequential events

recurrent themes of time and repeated events

No Comments

No Known Activations