INDEX

Explanations

words related to law, regulations, and authority

oai_token-act-pair · gpt-3.5-turbo

references to events or contexts that involve timing or specific years

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

cffffcc

-0.74

chieve

-0.70

ãĤ´ãĥ³

-0.66

oret

-0.64

ocaly

-0.64

phal

-0.63

undrum

-0.61

 Read

-0.61

 Starts

-0.60

eah

-0.60

POSITIVE LOGITS

 nobody

0.85

there

0.85

 they

0.84

 there

0.82

they

0.81

"[

0.78

 "âĢ¦

0.75

 none

0.73

 THEY

0.70

 attackers

0.66

Activations Density 0.295%

words related to law, regulations, and authority

references to events or contexts that involve timing or specific years

No Comments

No Known Activations