INDEX

Explanations

words related to a specific type of activity or event, particularly one involving participation and interaction

oai_token-act-pair · gpt-3.5-turbo

references to specific session-related contexts or activities

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

actor

-0.71

umbai

-0.64

kowski

-0.62

 soluble

-0.62

 traff

-0.61

tymology

-0.60

ging

-0.58

wolves

-0.58

warn

-0.57

 feud

-0.57

POSITIVE LOGITS

 session

0.95

ional

0.91

ansion

0.88

UAL

0.88

 sessions

0.85

ually

0.82

uality

0.82

 Session

0.79

ickr

0.77

eers

0.75

Activations Density 0.024%

words related to a specific type of activity or event, particularly one involving participation and interaction

references to specific session-related contexts or activities

No Comments

No Known Activations