INDEX

Explanations

mentions of the term "Med"

oai_token-act-pair · gpt-3.5-turbo

mentions of "Med" or variations thereof

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

chnology

-0.80

 innocence

-0.74

ngth

-0.69

holding

-0.65

 indign

-0.64

 Lauder

-0.62

 wart

-0.62

 Wilde

-0.61

pora

-0.60

 Macintosh

-0.60

POSITIVE LOGITS

iterranean

1.31

ications

1.13

itations

1.12

ication

1.11

itated

1.11

ient

1.11

itating

1.11

ieval

1.09

allion

1.09

icated

1.07

Activations Density 0.031%

mentions of the term "Med"

mentions of "Med" or variations thereof

No Comments

No Known Activations